Fighting Poverty with Index Insurance

Index insurance remains a promising new tool to help alleviate poverty by reducing the impacts of climate shocks in the developing world. It may even increase the poor's resiliency to climate change. In October, IRI co-hosted a workshop to discuss the technical challenges that currently preclude the use of index insurance on a large scale. We share here some of the experiences and resources to come out of that meeting.

"During the workshop, we learned about some of the scientific innovations that could help overcome the hurdles to scaling up insurance programs," says IRI's Molly Hellmuth, one of the event's organizers and editor of the Climate and Society Publication. "However, the innovations must be balanced with the reality on the ground: we need simple, understandable and trustworthy products if impoverished communities are to use index insurance successfully."

More than 30 experts from fields as diverse as reinsurance, climate science, economics and food security participated in the two-day workshop, which was co-hosted by the Center for Research on Environmental Decisions. Among them were representatives from the World Food Programme, the International Fund for Agricultural Development, the United Nations Development Programme and Oxfam America.

"As a development agency, we strive to support our national counterparts in broadening their options," says Mirey Atallah, who works in the UNDP's Environment Finance Group. "Index insurance is one of many options which we could potentially offer our 160 partner countries. It sends a financial signal to correct policies or practices that may not be viable in the face of climate change and it provides a financial contribution to other social safety nets."

Index insurance contracts are linked to a weather index such as rainfall, rather than a possible consequence of weather, such as crop failure. This subtle distinction resolves a number of fundamental problems that make traditional insurance unworkable in rural parts of developing countries. For example, unlike traditional crop insurance, the insurance company doesn't need to visit a farmer's field to determine premiums or to assess damages. Instead, the insurance contract is designed around rainfall data. If the amount of rainfall is below the threshold stipulated in the contract, the insurance pays out. Since the payout isn't linked to crop survival or failure, the farmer has incentive to make the best decisions for crop survival.

Case studies and pilot programs have shown that index insurance can effectively target a range of critical climate and poverty issues - from national-level food insecurity response to farm-level credit availability (read about IRI's work in Malawi and Ethiopia). These smaller scale projects show great promise but they also have uncovered significant questions which, if ignored during implementation on a larger scale, could lead to failure and even increase people's overall vulnerability.

Perhaps the largest hurdle to successfully scaling up index insurance is how to reduce and communicate basis risk. Basis risk is defined simply as all the risks that the insurance contract doesn't cover. Basis risk can occur for many reasons. It can occur when the amount of rainfall measured at a station differs from the rainfall at a farmer's plot. It can occur when the formula for the index doesn't fully reflect the drought risk or period when a crop is most vulnerable to drought, or if the index doesn't cover a useful risk in the first place. It can also occur if the crop is destroyed in a non-drought year by another factor, such as disease or flooding.

"People need to know what is not covered so they can protect themselves appropriately," says Dan Osgood, who leads IRI's research on index insurance. "The index we choose must of course target the correct risks, and whatever risk remains must be clearly understood by the contract holder."

Osgood says there's a tradeoff between providing complex insurance programs that have low basis risk and more simple ones that cover fewer risks.. "If the contract is too complex, people won't be able to use it or understand it properly, and can set themselves up for disaster.

Among the many other topics discussed during the meeting were how rainfall simulations and remote sensing could be used to design better insurance contracts and or communicate risks to farmers and other insurance users.

These technologies haven't been used on a broad scale for insurance programs that cover crop losses, mainly because of their limited resolution. For example, a remote-sensing product such as the Normalized Difference Vegetation Index (NDVI), an indicator of the amount and health of vegetation as measured by satellites, has a spatial resolution of four to eight kilometers.

"This is a problem for monitoring crops in regions where agricultural production comes predominantly from small holder farms, which are at spatial scales finer than one kilometer," says IRI scientist Pietro Ceccato. "On the other hand, in some instances even this low resolution is still superior to that of many ground-based rain-gauge networks in the developing world."

Another topic of discussion was the role of seasonal climate forecasts in the design of insurance programs. If the insurance is designed without accounting for forecasts but clients such as farmers use forecast information to buy insurance only in years when drought conditions are predicted, then the insurance financing may collapse.

The full selection of topic papers and presentations are available for download on the workshop's home page.

The workshop also represented an important step in the production of the next issue of IRI's flagship Climate and Society Publication, which will take a fresh look at the effectiveness of index insurance for reducing poverty and better managing climate risk. Climate and Society No. 2 will examine the current case studies, and rely on expert scientific opinion to delineate the advances, opportunities and pitfalls faced in scaling up index insurance.

"This next issue will hopefully help policy makers, researchers, donors and practitioners gain a better sense of what needs to be done to move forward," says Hellmuth.